Download Fish and Fisheries 2002 species concepts individuation

F I S H and F I S H E R I E S, 2002, 3, 171^196
On biological species, species concepts and individuation in
the natural world
Richard L Mayden
Department of Biology, Saint Louis University,3507 Laclede Avenue, St. Louis, MO 63103, USA
Abstract
I am a realist and argue that biological species exist in nature. I also argue that the
validity of ¢ndings of the many disciplines within the natural sciences employing
biological species in their endeavours of inquiry are unequivocally linked to the
accuracy of the species used in experiments. Few scientists today see the fundamental importance of taxonomic and systematic studies in both addressing accuracy of
diversity and the delineation of species diversity for other areas of science. The basic
controversial issues in the debate revolve around opinions regarding the nature of
species as either Individuals or Classes, confusion of Species as a taxonomic category
and as entities in nature, the varied practitioners studying diversity, a general lack
of a Lineage perspective and a gross chauvinistic perspective on the types of data
worthy of exposing and delineating diversity. I argue that species in nature are Individuals and form Lineages. As Individuals, they cannot be de¢ned but can only be
diagnosed in time. The category Species is a Class with a de¢nition. The di⁄culties
realised by scientists studying biodiversity in ‘de¢ning’ a species hinges upon the fact
that as natural entities they cannot be de¢ned. Recognizing and understanding the
origins of characters in species is further complicated if one views species in nature
as Classes and lacks an appreciation for the Lineage and the origin and retention of
traits through time. This forms an interesting paradox that many scientists have
fallen victim to wherein species are viewed as Classes (hence de¢nable, but immutable) yet are used to understand the process of descent that involves lineages and
Individuals! The pre-Darwinian Class perspective of species, combined with a common chauvinistic perspective on characters ultimately delays progress and places a
‘glass ceiling’ on species diversity for planet Earth. One resolution to the species
and species concept issue is to view the concepts in a hierarchical manner of primary (theoretical) and secondary (operational) concepts. Interestingly, the issue of
Individuals versus Classes for naturally occurring entities is much more widespread
and exists in many other scienti¢c ¢elds. Thus, a hierarchical perspective of having
a primary, nonoperational concept for natural entities and multiple operational concepts serving as ‘tools’ for discovering natural things consistent with the primary
concept is a heuristic methodology that is applicable to the advancement of many
areas of science.
Correspondence:
Richard L Mayden,
Department of
Biology, Saint Louis
University, 3507
Laclede Avenue, St.
Louis, MO 63103,
USA
Tel: þ1 314 977 3494
Fax: þ1 314 977 3658
E-mail: maydenrl@
slu.edu
Received 27 Feb 2002
Accepted10 June 2002
Keywords biodiversity, classes, conservation, species, species as individuals, systematics of ¢shes
# 2002 Blackwell Science Ltd
171
On species and species concepts R L Mayden
Introduction
172
What are species?
173
Some basics and known caveats
175
What are individuals and what is individuation?
178
Alternatives to individuals?
180
Individuals versus classes: what’s the big deal?
180
Why then are Individuals so important to clearly identify?
183
The hierarchy of concepts of species
184
A parallelism of monophyly and the evolutionary species concept
184
Data chauvinism, aesthetics and evidence for the existence of a species
186
Implications of species concepts on the diversity of fishes
189
A glass ceiling on species diversity
190
Returning to the hierarchy of concepts
191
Individuals above and below the scale of species
192
Acknowledgements
193
References
193
Introduction
The nature of biological species has seduced scientists from many disciplines for centuries and has
resulted in an overwhelming amount of literature
and debate. Given this intense focus, it would necessarily follow that the topic of species must be something of major signi¢cance in science. This is
exempli¢ed by the continued number of papers
addressing this complex, yet fundamental, issue and
the occurrence of three recent volumes speci¢cally
targeting the topic (Howard and Berlocher1998; Wilson 1999; Wheeler and Meier 2000). As commonly
perceived units of biodiversity participating in natural processes and the focus of many research paradigms, their characterization is fundamental to the
natural sciences. Although scienti¢c debate continues over what constitutes a species and di¡erent
conceptualizations of species, two fundamental
observations hold true. First, it is true that species,
as biological diversity, are essential elements of scienti¢c investigations for multiple disciplines, are studied at varied levels of scale and for an in¢nite
number of questions. Species are the basic currency
of biodiversity worldwide. Second, diversity is disappearing worldwide at unprecedented rates, and
international, national or state protection is largely
administered to biological entities that are referred to as species. Unfortunately, the consternation
172
surrounding the issue of species and species concepts has driven many practising scientists studying
biodiversity, population genetics, ecology and systematics away from the topic and in search of the
comfort of ‘safe’ or less controversial issues in their
disciplines. In fact, it is not uncommon to talk with
these scientists, educators, researchers and even policy makers and discover that they either have no
interest in the topic, have not followed the debate, do
not think that the topic pertains to their science, or
are simply confused by much of the philosophical jargon that is traditionally tossed about by those scientists and philosophers of science participating in the
exchange. This, of course, is detrimental to advancing multiple ¢elds of the natural sciences because
biological units (species) being used in investigations
may not re£ect legitimate by-products of descent,
thus calling to question the results of such studies.
The disjunction in both the scienti¢c and nonscienti¢c communities between theoretical and empirical
discussions, literature and research regarding species
is particularly disconcerting.Why? Because in these
instances data, empirical observations and discussions are interpreted or conducted within a theoretical, philosophical, or metaphysical vacuum. The
interpretation of raw data, and their presumed
patterns or relationships is done by individuals processing a particular worldview rooted in some preconceived theoretical framework that is traceable to
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
their academic heritage, experience and breadth of
knowledge of any ¢eld. However, this framework is
not clearly stated and obvious to a reader, thus preventing critiques, debates and advancement of
research hypotheses. Scienti¢c debate is good; from
these exchanges come advancement, though it may
come at a cost. In the case of biological species, the
extended healthy debate over what constitutes a
species, comes with a cost to both biodiversity and
individual scientists. Without resolution, biological
diversity unaccounted for by particular species concepts may go unnoticed and eventually go extinct.
For the individual scientist, the cost is time-necessary to stay abreast of this fundamental species issue.
As such, value judgements, and as we will see below,
aesthetics are also a part of the equation in dealing
with the issue of biological species.
As myself and others have outlined in previous
papers, the di⁄culty in resolving the many issues
surrounding biological species can be fundamentally
linked to the nature of their ‘existence’, as Individuals1, and the inherent di⁄culty in dealing with
things that are Individuals. Although in the natural
sciences, the discussion of biological species has
received considerable negative attention that has
been derisively summarized as ‘a bunch of scientists
and philosophers arguing over how many angels
can dance on the head of a pin’, the underlying reasons for the controversy in this issue is not unique to
the topic of biological species, but extends beyond
this topic, and is signi¢cant to numerous scienti¢c
disciplines including biodiversity. As I discuss in this
essay, understanding the nature of being an Individual is critically important in structuring our
thought, analyses and interpretation of empirical
data. The same issues surrounding Individuality of
biological species are common in other disciplines of
the natural, social and physical sciences but, interestingly, have not been the focus of such debate. It is possible that dialogues in these other disciplines do not
have the historical inertia as the biological species
issue or the scientists may be less aware of the important philosophical issues related to Individuality
underlying their science. However, as promulgated
herein, understanding Individuals and incorporating
this understanding into disciplines beyond biological
diversity will unquestionably provide for both a better education of prepared scientists to advance their
¢elds and, for those in biodiversity-related ¢elds, will
improve our success in acquiring important empirical data and properly interpreting these data in our
quest to account for, understand and conserve our
natural world.
What are species?
1
The term Individual is capitalized and in italics to denote my
reference to specific types of entities discussed by scientists
and philosophers of science (Ghiselin 1966, 1974, 1989,
1997; Hull 1976) and is explained in detail below. A particular
species in nature is an Individual; every organism, existing as
clonal bodies, products of sexual reproduction, or products of
symbiotic relationships are Individuals.
If di¡erent biological professionals and higher education students majoring in the sciences of the postDarwinian age were polled on this question, it is likely
that it would result in fewer answers than one might
expect. Without a doubt, the vast majority of the
introductory biology textbooks used in secondary
schools, colleges and universities provide limited discussion on the topic and are likely focused on biological species being reproductively isolated from other
species ^ the traditional Biological Species Concept
(BSC). Logically, in the disciplines closest to the species question there will be a much greater diversity
of opinions on what a species is. Aside from systematists and some philosophers of science, very few will
be familiar with the many alternative conceptualizations of species. However, very few, even those close
to the subject, will answer this question: ‘They are
scienti¢c hypotheses regarding the existence of a
unique and distinct biological and evolutionary
entity. They are hypotheses presented on the basis of
evidence that lead skilled researchers in systematics
and taxonomy to propose that some populations are
unique and form independent lineages relative to
other populations traditionally grouped with them.’
Thus, people describing species are conducting
scienti¢c research and presenting hypotheses about
unique and distinct biological entities that are the
result of and participate in the evolutionary process
as distinct lineages.
As part of the scienti¢c process, however, and as
hypotheses they need to be testable, and just like
other hypotheses generated in the natural sciences,
they can never be proven.Yes, hypotheses of this nature can be tested, although some may argue otherwise.When someone presents a hypothesis of a new
species, it is done in the context of the species concept
that the researcher adheres to at the time of the
description. What is the concept that the author
adheres to? How often does the author provide the
readers with the species concept that she/he is working under? Does the process of falsi¢cation of a given
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
173
On species and species concepts R L Mayden
hypothesis have to be done in the context of the given
species concept? Or can it be done within the context
of any species concept? Then what is the hypothesis
to be tested? Is it the species concept that is being
tested, that the concept appropriately recognizes species? Or is it the species itself? It is clear that there is
very little in the literature dealing with this question,
but it should be a question in the minds of scientists
conducting systematic, taxonomic and biodiversity
research. The answer to this question (discussed
below), I think, provides very signi¢cant insight into
resolving the species question. However, one of the
major points is that species descriptions represent
testable hypotheses, just the same as what is presented in other areas of the natural sciences.
For some scientists, genera, families, orders and
other supraspeci¢c taxonomic groups of the Linnaean hierarchy are thought of as ‘real’ things existing in nature. In other words, there are such things
that can be identi¢ed with generic, familial or ordinal
characters; should one ¢nd something possessing
one of these traits, then one has discovered one of
these supraspeci¢c taxa (Winston 1999). Education
and practice of this type of system was largely
conducted through close mentoring relationships
wherein ideas as to ‘what species, genera, families,
etc. are’ is learned ostensibly by the student interacting with the mentor. Traditionally, prior to phylogenetic systematics and the criterion of monophyly,
earlier systematists argued for new genera or families
because the taxa di¡ered from other such taxa by a
certain magnitude for characters, usually morphological. In this framework, genera were less di¡erent
from one another than were families and so on; some
even argue that some taxonomic groups should
include a given number of species. Thus, taxonomists
largely classi¢ed on the basis of di¡erences and the
magnitude of these di¡erences. Today, the systematic
community argues that supraspeci¢c taxa do not
exist in nature; they are manifestations of the historic
past through ancestor^descendant relationships
and are given proper names that we, as systematists,
superimpose on a phylogenetic tree to identify monophyletic groups ^ nothing more and nothing less.
These names serve as convenient communication
devices to relay information regarding groups of species at di¡erent hierarchical levels.
Are species like this? After all, all monophyletic
groups (supraspeci¢c taxa) begin as an individual
species. Are the things that we identify with a binomial constructs based in a human need to classify,
much the same as with supraspeci¢c taxa? Are there
174
such things as species characters that we are or
should be using to look for species? As such, if we
can have the sense of a species character does it
empower us in the process of discovering biological
diversity? Answers to these questions will depend
upon the person, the time in which the question is
asked, and even the group of organisms that a person
works with, if any.While some of these questions are
addressed below under di¡erent sections, it is worthy
of some discourse here.
For most modern systematists, species are not
like supraspeci¢c taxa. Supraspeci¢c taxa in the
domain of phylogenetic systematics are monophyletic groups. Monophyletic groups are natural groups
and contain the hypothesized or real common ancestor and all of its descendants (species). Monophyletic
groups only have historical cohesion as a result of
descending from a common ancestor; the group as a
whole does not participate in any natural processes
as a whole. There are no ‘species characters’ that one
can use to ¢nd species. A particular species di¡ers
from a particular supraspeci¢c taxon in that species
participate in the natural processes of selection, descent, modi¢cation, etc. all processes that a supraspeci¢c taxon cannot. That is, particular supraspeci¢c
taxa cannot, as a whole, participate in selection, descent and modi¢cation. Because supraspeci¢c taxa,
by de¢nition, consist of two or more species with a
unique common ancestor, following the ¢rst speciation event any type of cohesion (other than historic)
is lost. Species are thought to be entities in nature
that arise via speciation events; the exact nature of
these events is still largely unknown. Species are not
thought of as entities that are arti¢cial groupings
with no common history of descent, as would be the
case if they were created and grouped on the basis of
de¢ning characters. As descendants of a process of
character modi¢cation, any number of attributes in
the lineage can change and the lineage (species) will
defy de¢nition. It is impossible to predict the type
and degree of character modi¢cation that will result
from a speciation event. Thus, ‘species characters’
naturally do not exist any more than do characters
for genera, families, orders, etc.
The comment is often made that ‘sure, species
may be elements of evolution but evolution operates
on smaller groupings like populations and metapopulations.’ Thus, for many there is a concern over
whether species are the units of evolution or are the
smaller groups of individuals organisms the units of
evolution? From my perspective, the answer to this
question is yes and yes. All of these groupings of
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
phies of the supraspeci¢c taxon that they form the
root (Wiley 1981). Given this twist, is it not odd that
phylogenies are reconstructed routinely using various types of data and that monophyletic groups (or
common ancestral species) are defended on di¡ering
numbers of synapomorphies or probabilities, but
recognition of valid, extant descendent species as
fundamental units of biodiversity is held to a di¡erent
criterion? This paradox, the nature of species as Individuals (not Classes2), and the nature of humans
gravitating towards chauvinistic behaviours and
aesthetic priorities in character choices for species
forms the body of the following dialog on species.
individual organisms make up entities that participate in evolutionary processes, and the exact make
up of these entities can and will vary over time. However, biological species are perceived as the highest
level of organization wherein tokogenetic relationships between individuals are maintained for cohesion and identity (Wiley and Mayden 2000a,b,c).
Species are also perceived as those things that maintain themselves as independent lineages and have
cohesion, and change over time through modi¢cations during descent. If this is not the case, then the
whole underlying principles of evolution are problematic. A species undergoing a speciation event
changes from one Individual with tokogenetic relationships to two Individuals having phylogenetic relationships. This leads to questions regarding the
di¡ering population structures that may exist within
a species. How does one know to identify something
as a species and christen it with a proper name as an
independent entity of evolution? Where is the natural
cut-o¡ between things that are clearly populations
or metapopulations and things that are species?
These questions are as di⁄cult to address operationally as questions regarding the process of speciation
and how it happens ^ because they are one and the
same.The di⁄culty for humans to clearlyanswer this
question relates to the fact that, as a species, we have
a tendency to assign values to things, view many
things from an aesthetic perspective, and desire
clear boundaries. However, if we strip these unquanti¢able assessments from our decisions, then
we are left with a basic underlying principle of descent with modi¢cation, that is, species are lineages that maintain their independence from other
lineages and participate in natural processes. Only
when we impose or incorporate values and aesthetics
does the question arise ^ what types of characters
should be used to identify species and what degree
of di¡erence is needed for an entity to qualify as a
species?
One remaining point that often escapes the
thought process of many is that particular species
have historically participated in the speciation process that lead to the production of two or more species
(sister species), and, as such all natural, monophyletic
supraspeci¢c taxa begin as a single species, no di¡erent
than species that neotological scientists research
today in divergent disciplines. Hence, ancestral species are those taxa that gave rise to other taxa
through speciation, and the autapomorphic traits
that they possessed, be they morphological, molecular, behavioural or ecological, are the synapomor-
The term Class is capitalized and in italics to denote my reference to specific types of groupings of entities discussed by
scientists and philosophers of science (Ghiselin 1966, 1974,
1989, 1997; Hull 1976) and is explained in detail below. A particular species is an Individual in the Class construct of the
Linnaean Hierarchical System Category Species. Likewise, a
single helium atom is an Individual in the Class Helium that
we visualize in the periodic table.
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
175
Some basics and known caveats
Any discussion on biological species should be prefaced with a brief discussion of a number of points
that are basic to the issue and admonitions of the current state of this discipline. Herein, I outline some
key issues that readers must keep in mind about our
current state of knowledge, the real emphasis on progress in understanding species, requisite information
needed to fully achieve our quest for species, why so
many di¡erent conceptualizations of species are in
existence today, and the existence of personal and
social bias in this ‘scienti¢c’process.
While the conception of species has changed, from
historic time scientists have recognized biological
species as fundamental elements in the natural
sciences. Even prior to the Darwinian revolution,
when descent with modi¢cation replaced previous
theories explicating the diversity of life, researchers
identi¢ed species as entities participating in processes and having essential attributes. Theories and
concepts are fundamental links between pattern
and process in the natural world, and are employed
by everyone, not just scientists, to aid in our understanding and interpretation of the world around us.
They derive from our metaphysical world view, provide a framework in£uencing our observations and
the kinds of observations made, and serve as fundamental bridges between observations to interpretations and conclusions. Today, most scientists work
2
On species and species concepts R L Mayden
systematic biology to recover patterns of ancestor^
descendant relationships, or common ancestral species. As unique homologueous modi¢cations, they
are the clues used to detect the unique history of descent. They are modi¢cations that occurred in previously existing species. The types of characters or
traits used in systematic biology to recover evolutionary histories are extremely varied, including such
traditional attributes associated with morphology to
behaviour or ecology, to many genetic markers
including protein, DNA and RNA sequences, cytology, and 28 and 38 structures of molecules. Given
the numerous comparative systematic studies that
have been conducted on varied organisms, using various types of characters, and at various taxonomic
levels, two things have become abundantly clear.
First, not all types of traits (morphology, genetics,
behaviour, etc.) change at a constant rate within and
between taxa. Second, not all types of traits change
at the same rate within and between taxa. Logically,
these two empirical observations should make one
take notice that not only does this endanger any
notions of a ‘clock hypothesis’ for particular characters, but that di¡erent types of traits are important
in resolving the tree of life and in identifying species,
either as ancestors or descendants!
The species concept issue has been a continuing
saga of controversy for many reasons, all equally justi¢able to researchers proposing them, and has
resulted in many, many publications focusing on the
topic. As a result of years of controversy and antagonism surrounding this debate, there have been over
20 di¡erent conceptualizations about biological species, what constitutes a biological species, and what
does not (Mayden 1997,1999; seeTable 1). All, but the
Evolutionary Species Concept (ESC) (Mayden and
Wood 1995; Mayden 1997, 1999; Wiley and Mayden
2000a,b,c; Coleman and Wiley 2001; Stau¡er and
McKaye 2001), are highly operational and serve as
convenient, operational de¢nitions of things known
as Classes or Natural Kinds3, that is, nearly each concept provides the researcher with a clearly outlined
criterion for when something is or is not a valid biological species. In the BSC, the taxa need to be sexually
reproducing sister species (derived from a unique
within the metaphysical paradigm of descent with
modi¢cation, not pre-Darwinian paradigms invoking
the creation of diversity without diverging lineages
underlying the speciation and diversity observed
today. Thus, diversity results from some degree of
selective process, divergence, lineages through time
and speciation.
Directly related to this notion and addressed to
some degree below is the fact that a dialogue on biological species almost always involves theoretical and
operational issues, and these issues are often con£ated. Discussions on concepts of species often really
focus on the operational issues of the recognition of
such entities and argue for particular views making
the identi¢cation process more clearly operational.
Thus, by focusing on such discussions, species will
be those things that favoured concepts will recognize
in the operational process (sexually isolated, morphologically distinct, etc.). Operational concerns, like
‘Is this a reasonable way to diagnose species?’ or ‘Is
10% divergence (or some other arti¢cially derived
number) for a particular gene indicative of a valid
species?’or ‘Exactly how much reproductive isolation
is prerequisite for speciation?’ have their place in discussions of practice. However, the theory and metaphysics that should bear on and underlie questions
such as these and the general science of evolution,
diversi¢cation, or biodiversity are often given lesser
concern, are disregarded, or may be supplanted by
some apprehension for discrete limits in de¢ning species. Complete focus on operational concerns, without reference to theory, leads to a process that is not
scienti¢c.
With the acceptance of descent with modi¢cation,
the scienti¢c community adopts a number of underlying premises for theoretical concepts. These
include: (i) selection and other factors that operate
on lineages, (ii) lineages change over time via anagenesis, (iii) modi¢cations during descent occur with
any number of types of attributes, (iv) heritable modi¢cations are passed on to descendants in lineages,
and (v) lineages, at some time, cease to exist as single
entities via speciation (or extinction). These are basic
underlying postulates in their simplest terms, of comparative evolutionary and systematic biology. If the
whole process of descent with modi¢cation via
lineages does not hold true then much of our search
for diversity, establishing evolutionary relationships
of this diversity, and understanding the evolutionary
mechanisms behind diversi¢cation are nonsense.
With respect to anagenesis, the changes that occur
in lineages over time are the very attributes used in
Natural Kinds are special types of Classes that include as
their members entities that participate in natural governing
processes and theories. A helium atom is in the Natural Kind
Helium and is subject to the laws of atomic theory; Earth is
in the Natural Kind Planets and is subjected to the laws of
planetary theory.
176
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
3
On species and species concepts R L Mayden
Table 1 Various species concepts,
abbreviations used in text and
authors promulgating individual
concepts.
Agamospecies concept (ASC)a
Biological Species Concept (BSC)b
Cladistic Species Concept (ClSC)c
Cohesion Species Concept (CSC)d
Composite Species Concept (CpCS)e
Ecological Species Concept (EcSC)f
Evolutionary Significant Unit (ESU)g
Evolutionary Species Concept (ESC)h
Genealogical Concordance Concept
(GCC)i
Genetic Species Concept (GSC)j
Genotypic Cluster Definition (GCD)k
Hennigian Species Concept (HSC)l
Internodal Species Concept (ISC)m
Morphological Species Concept (MSC)n
Nondimensional Species Concept (NdSC)o
Phenetic Species Concept (PhSC)p
Phylogenetic Species Concepts (PSC):
Diagnosable Version (PSC1)q
Monophyly Version (PSC2)r
Diagnosable/Monophyly Version (PSC3)s
Polythetic Species Concept (PtSC)t
Recognition Species Concept (RSC)u
Reproductive Competition Concept (RCC)v
Successional Species Concept (SSC)w
Taxonomic Species Concept (TSC)x
Concepts and abbreviations are from Mayden (1997). Superscripts correspond to authors
employing or describing the concept. All concepts are reviewed in Mayden (1997).
a
Stuessy (1990); bMayr (1940, 1957), Mayr and Ashlock (1991), Mayden and Wood
(1995); cRidley (1989); dTempleton (1989); eKornet (1993), Kornet and McAllister (1993);
f
Van Valen (1976); gWaples (1991, 1995), Mayden and Wood (1995); hWiley (1978), Frost
and Hillis (1990), Mayden and Wood (1995), Wiley and Mayden (2000a); iAvise and Ball
(1990); jSimpson (1943), Dobzhansky (1950), Mayr (1969); kMallet (1995); lHennig (1950,
1966), Meier and Willmann (2000); mKornet (1993); nCronquist (1978), Shull (1923), Du
Rietz (1930), Regan (1926); ovaried concepts that lack a lineage perspective to
interpreting the origins and evolution of characteristics, including reproductive isolation;
p
Sneath (1976); qEldredge and Cracraft 1980), Cracraft (1983), Nixon and Wheeler
(1990), Wheeler and Platnick 2000); rRosen (1978, 1979); sMcKitrick and Zink (1988);
t
various concepts that employ a combination of characteristics to diagnose or define
species with no temporal or lineage perspect to the evolution of species and their
attributes; uPaterson (1993); vGhiselin (1974); wpalaeospecies concept of Simpson 1961)
and chronospecies concept of George (1956); xBlackwelder (1967) (modified from
Mayden 1999).
common ancestor) in sympatry and reproductively
isolated from one another. Do asexual species have
any ontological status? How would asexual species
be recognized? Given that the vast majority of speciation is allopatric and sister species do not occur
together (Lynch 1989; Grady and LeGrande 1992;
Chesser and Zink 1994), how can the species recognized under this concept that live in allopatry be
tested? This concept, if employed by its operational
guidelines, would account for only a trivial amount
of Earth’s biodiversity. In one version of the Phylogenetic Species Concept (PSC), species are only those
things that possess autapomorphic traits (Table 1).
What about ancestral species? Ancestral species, by
de¢nition, do not possess any autapomorphic traits;
these species can be discovered through thorough
analyses but only possess the synapomorphic traits
of the clade. They will be involved in a polytomy with
other members of the clade to which they gave rise,
and will lack any uniquely derived traits. In the
Genetic Species Concept, species are only those
things that di¡er from one another by a set genetic
distance.Who determines the genes or proteins to be
examined? Or, better yet, who gets to determine the
magic distance at which species diverge? Finally, in
the Ecological Species Concept species are those
things that have divergent ecologies; thus, can no
two species in sympatry or allopatry possess the same
ecologies? Given that the vast majority of speciation
occurs in allopatry would one expect all speciation
events to be accompanied by ecological changes? I
suspect not, given that close relatives and potential
‘competitors’ would live in allopatry. Many ecologies
of species living in allopatry will likely be very similar, if not the same, because the species are derived
from unique shared common ancestors. Given the
tremendous number of constraints of the BSC, this
concept would also only account for a trivial amount
of diversity, and is likely the worst concept that any
area of biology should adopt.
Controversy surrounds the topic of species, lineages and the discovery of both. Why? It is related to
several key components that are both sociologically
and scienti¢cally based, and issues that have yet to
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
177
On species and species concepts R L Mayden
be determined. First, it is the nature of species being
Individuals (discussed below) and, as such, they will
always have fuzzy boundaries on them. Second,
scientists have diverse academic backgrounds and
they adopt perceptions of reality and judgements
re£ective of their backgrounds. If scientists have a
more ecological or genetic background in their academic history, then they are more likely to be swayed
by or employ concepts emphasizing these qualities.
If one is a morphological taxonomist or systematist,
then there will likely be more emphasis placed on
these types of traits validating the existence of a species, even in the face of dramatic genetic di¡erences
between morphologically similar ‘forms’. Sociologically, current-day practitioners of natural sciences
related to the recovery of biodiversity are heavily
swayed by the advertised potential to discover species as part of biological diversity using molecular
tools. Although the methods being employed in
molecular studies to unearth natural variation can
be more informative in some cases, it is equally likely
that they will either not provide information in other
cases or the information generated is of questionable
validity to ask evolutionary and biodiversity questions because of questions of homology (Stau¡er and
McKaye 2001). Yes, molecular methods and studies
do indeed have problems in data recovery and analysis to the same degree as traditional morphological
studies or those studies using protein gel electrophoresis!
Another area of critical importance that has
silently permeated the ‘species concept issue’ has
been the rules of nomenclature. These rules seek stability, advocate a type concept, minimize emphasis
on variability, and are ¢ne for a world where nothing
changes over time (e.g. lineages) and the entities in
the world are Classes or Natural Kinds (see below). To
be fair, as we will see below, the world does consist of
a number of Natural Kinds or Class constructs and
nomenclature rules work quite well for these things;
however, species are not these things. At the level of
species and below entities are changing over time,
di¡er in space, and do not ¢t the general ideas promulgated in the nomenclatorial documents. Thus,
researchers need to be cognizant of this paradox and
realise that the rules do exist for naming but should
not re£ect our metaphysical view of the world today.
Finally, no candid discussion of species concepts
should go without addressing some admonitions of
the ¢eld and those working in the ¢eld. First, clearly
there remain many types of biological ‘entities’ that
we really don’t know what to call ^ especially in the
178
world of unisexual, asexual, or parasexual organisms. Those working with sexually reproducing
organisms have conducted the vast majority of the
‘species’ work, and most of the targeted groups are
vertebrates. Sociologically, this is where most researchers feel most comfortable in their investigations; lineage cohesion, formation, maintenance
and splitting in a world without sex is perplexing
and somewhat abstract, and can actually lead to
some very interesting and bizarre discourse as to biological diversity. Thus, we really don’t have it all covered in this saga and many more papers will develop
on this subject. This must be realised in our continuing studies of the‘Species Quest’, and those searching
to succeed in completing our understanding must
be cognizant of the fundamental signi¢cance of
metaphysics, theory and philosophy of science, and
the appropriateness and limitations of empirical evidence. Likewise, as also outlined below, researchers
must also be aware of the inherent bias of our traditional attempts to discover and understand species
because of human-induced value systems and aesthetic considerations.
What are individuals and what is individuation?
The Individual, in a philosophical sense, has been discussed in considerable detail by philosophers and
systematists Ghiselin (1966, 1974, 1989, 1997), Hull
(1976), Frost and Kluge (1994), Mayden (1999), Wiley
(1981), Coleman and Wiley (2001) and Stau¡er and
McKaye (2001), to mention only a few. As outlined in
Fig. 1(A), Individuals are entities with restricted spatio-temporal frameworks and have both cohesion
and continuity. They are capable of self-replication,
are particulars with a unique beginning and ending,
and are parts of a whole. Individuals can change over
time and space, so they may only be diagnosed and
described, but not de¢ned.Theycan lose some of their
parts or gain additional parts and still remain the
same Individual, depending upon what is and how
much is lost or gained. They participate in natural
processes as evidenced by their cohesion and continuity over time and under various selective or nonselective regimes. Because Individuals cannot be
de¢ned, the desired clear-cut boundaries are inherently impossible, and they are often referred to as
‘fuzzy.’ If an entity is an Individual, because of these
boundary issues, one will often see multiple pragmatic or operational de¢nitions for the entity. It is
also true that there will be disagreement among
those posing de¢nitions for the entity. Not every
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
Figure 1 A simple tabular and
conceptual diagram comparison of
Individuals and Classes. The
characters being compared for both
Individuals and Classes are identical.
Graphic representations illustrate the
fundamental di¡erences in
recognizing species as Individuals and
Classes.
researcher will focus on the same attributes of the
Individual in their de¢nitions because each will see
di¡erent things to qualify, quantify and describe, and
will likely have their de¢nitions in£uenced by their
own values and aesthetic priorities.
In the example of an Individual (Fig. 1A), I illustrate
a single lineage undergoing lineage-splitting via speciation, divergence via anagenesis, and the termination of an Individual via extinction. Prior to the
speciation event a single lineage maintained cohesion and continuity since its unique origin. During
the speciation event, two new Individuals were
derived (or the ancestral lineage could survive and
one new species is derived) with each having independence, cohesion, a new beginning, and an eventual end. Each lineage is constrained by the
properties that it inherits from its ancestral lineage
and those resulting from the speciation event, plus
any unique traits that evolve after the speciation
event. All of these attributes will, at some time, eventually serve as historical constraints in development,
ecology, physiology, genetics, etc. This process of generating new Individuals, as well as the process
through which we identify Individuals, I refer to as
Individuation.
Biological species are Individuals and the process of
their formation is speciation, a type of Individuation.
Individuals are entities in the natural world that we
hope to discover, investigate, use in our experimental
studies, use to test hypotheses or theories, use for
developing predictions, and they are the fundamental observations in the foundation of theories of the
natural world. As species, they are subject to selective
pressures and anagenesis. Thus, one might expect
that if Individuals are subject to natural forces and
can change in time and space, lack clear boundaries,
and cannot be de¢ned, then they are likely to be di⁄cult to delineate and will come with some level of
scienti¢c uncertainty. Depending upon the particular criterion that a researcher may focus on to delineate an Individual (morphology, genetics, etc.), the
boundaries may di¡er from one researcher to
another and from one criterion (character) to
another. Yes, the same Individual may have di¡erent
hypothesized boundaries because of di¡erent rates
of change in or di¡erential emphasis on each of the
criteria. In fact, unless all criteria for delineating an
Individual are all changing in the Individual at exactly
the same rate temporally and spatially, then there
will appear to be inconsistency in the delineation of
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
179
On species and species concepts R L Mayden
Table 2 Examples of Individuals and Classes.
Individual
Class or Natural Kind
Hydrogen
Mesenchyme cell
Rick Mayden
Species (in nature)
Monophyletic group
Horseshoe Lake, IL, USA
Earth
Milky Way
Hydrogen
Cell
Organism
Linnaean species category
Linnaean supraspecific category
Ecosystem
Planet
Galaxy
Individuals on the left can be placed in the Class or Natural Kinds
on the right. See Fig. 1 for the basic comparisons of qualities of
Individuals and Classes.
the Individual. This inconsistency only exists, however, if one lacks an appreciation for the qualities of
the Individual and the need to evaluate cases such as
this with a temporal perspective. Even with di¡ering
rates of change in time and space, if the Individual is
viewed from a historical perspective (phylogenetic
systematics), one can make sense of the di¡ering
boundaries based on the di¡erent criteria for existence. Some examples of Individuals are provided in
Table 2. The organism Rick Mayden is an Individual
and it can be diagnosed at various stages of its life
transformation. However, if di¡erent researchers
each decide on the delineation of Rick Mayden and
develop this at di¡erent times in his ontogeny, with
or without di¡erent criteria, then there will be di¡ering and con£icting de¢nitions of Rick Mayden.
Alternatives to individuals?
If species are not Individuals, then what are they?
Some prefer to think of species as Class constructs,
either as arti¢cial Classes or as Natural Kinds (Kitts
and Kitts 1979; Kitcher 1984, 1987, 1989). First, the
term Class should not be confused with the taxonomic category of the Linnaean hierarchy termed
Class (nonitalic Class refers to the taxonomic category). Classes and Natural Kinds (Fig. 1B) are
unbounded spatiotemporal constructs with members, and membership is determined by the de¢nition
of the Class. They are not self-replicating. Any entity
that ¢ts the de¢nition belongs to the Class regardless
of its historic origin or location in time and space.
Classes do not participate in processes as a whole,
they do not change in time and space, and hence, they
have very clear and infallible de¢nitions. Thus, members of a Class can have origins in di¡erent locations
180
on Earth or anywhere in the Milky Way Galaxy and
they are immutable (prerequisite to staying in the
Class), but the Class itself never has a beginning or
end and does not participate in any process. Some
examples of Classes are provided in Table 2. If the
organism Rick Mayden is a Class, then the ontogeny
of Rick Mayden requires an in¢nite number of de¢nitions for the in¢nite number of stages of his life transformation. The same is true of individual species; if
species are Classes, then each requires an in¢nite
number of de¢nitions to accommodate an in¢nite
number of molecular, morphological, physiological,
behavioural, etc. changes occurring in the descent
of each lineage. If species are Classes, they are immutable and only require one de¢nition; consequently,
descent with modi¢cation cannot exist if species are
Classes.
Individuals versus classes: what’s the big deal?
One of the most important issues facing the natural
sciences today and delaying important progress
toward understanding, and compromising our scienti¢c progress in developing appropriate and interrelated ¢elds in biodiversity, evolution, ecology,
systematics, and conservation biology, is a continuing tradition of scientists thinking of and treating species in nature as Classes or Class constructs, that is,
many natural scientists continue to think about and
treat species like essentialists, ¢ctionists, phenomenologists, or nominalists ^ not realists. Entities like biological species exist in the real world and can be
discovered in nature and used to investigate natural
processes. They are not ‘phenomena’or ‘arti¢cial constructs’ that we recognize and arrange in some order
perceived in the mind of a particular scientist.
This is an extremely important issue to consider
because it creates an interesting paradox: scientists
seek ‘de¢nitions’of every particular species as if they
are immutable, even given our currently accepted
metaphysical worldview regarding the origin of the
diversity of life. The paradox is simple. How can one
view or treat particular species in nature as Classes
or Natural Kinds and yet aspire to investigate real processes associated with their origins and their continued existence?
The history of scienti¢c thought and the scienti¢c
process has been clearly marked with prominent
metaphysical and paradigm shifts (Kuhn 1962). Theories and concepts are formulated and embodied in
a particular metaphysical worldview of the time of
their development. Ideas, hypotheses, causative
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
explanations, and even observations of reality all
derive from the metaphysical paradigm or theory in
existence at that time. Every discipline can identify
metaphysical shifts in their timeline that occurred
for a variety of reasons, including the acquisition of
new knowledge, mergers of disciplines, improved
instrumentation, new primary theories exposed to
strong scrutiny and rejected, etc. For example, one
may examine the history of our inquiry of the cosmos
and identify a series of important metaphysical
shifts extending from the Ptolemaic (Earth-centred)
Model of the Heavens, to the Copernican Helicocentric (Sun-centred) Cosmos, to the Descartes Model
embodying a mechanical philosophy, to the Keplerian view of elliptical motion. In all of these changes,
a new worldview came about that altered how people
interpreted the world around them, impacted how
data are interpreted, and provided constraints on
how information is processed by individual scientists
and nonscientists.
Probably, one of the most signi¢cant of the metaphysical shifts that has ever occurred in the sciences
involved the explanation for the origin of biological
diversity. The change in the scienti¢c worldview as
to the origin of biological diversity from that of immutable entities formed by a Creator to entities existing
because of evolutionary descent through natural
selection and modi¢cation was clearly a signi¢cant
metaphysical shift in the natural sciences. From the
Darwinian paradigm of lineages existing through
time and consisting of ancestral and descendent species evolved the ideas of speciation, multiple types of
selection and speciation, and many other areas of
evolutionary biology. Consistent with all of these
endeavors of evolutionary biology are the notions of
lineages, descent, stasis, anagenesis, and heritable
attributes, to mention just a few.
Paradoxically, none of these notions are even tenable, and the theory of evolution is completely
implausible, unless one views species as Individuals
and not Classes or Natural Kinds. As the former, species are natural entities, lineages, to be discovered,
but they lack de¢nitions and can only be diagnosed
retrospectively (Frost and Kluge 1994). They are
inherently di⁄cult to de¢ne, but as lineages they
can be characterized or diagnosed on the basis of
any heritable attributes that provide evidence of lineage independence. Likewise, as Individuals, they
change over time and it is impossible to predict what
attributes will be changing in their descent (e.g. morphology, genetics, behaviour, etc.). Thus, in order to
discover these natural by-products of descent
researchers must search for evidence of lineage independence. Practitioners of science requiring particular de¢nitions or a degree of di¡erentiation for
particular qualities (outlined in various species concepts) be met before a species is considered valid view
and will treat biological species as immutable Classes
or Natural Kinds, a metaphysical worldview held
prior to the Darwinian revolution. All of the current
concepts of species except for the ESC treat biological
species as Classes or Natural Kinds (Mayden 1997,
1999). Thus, it is a di⁄cult position to be in as a contemporary evolutionary biologist, systematist, geneticist, or any other type of biologist advocating
descent with modi¢cation and at the same time possess a metaphysical world view of species as immutable kinds as was thought prior to the Darwinian
revolution.
Only the ESC is acceptable as a guide to discovering
diversity referable to biological species as Individuals;
all other concepts require a predetermined type of
divergence or process. The search for clearly de¢nable products of nature meeting predetermined standards inherent in operational concepts or de¢nitions
formulated because of pragmatic concerns or arguments clearly identi¢es a scienti¢c culture satis¢ed
with species as Classes or Natural Kinds. Such concepts lack the breadth of temporal and spatial concepts to interpret the origins of attributes possessed
by species, deny the existence of some species diversity diagnosable on the basis of other types of traits,
require the existence of some special qualities before
lineages are recognisable, and among a number of
other malformations, preclude the study of evolutionary processes (Mayden1997,1999;Wiley and Mayden
2000a,b,c). These concepts not only provide de¢nitions or operational tools for discovering only certain
types of diversity consistent with the concepts in
favour at a particular point in time, but they likewise
serve to ful¢l predetermined statements of value
and/or aesthetic priorities for particular kinds of species and traits acceptable to recognize such diversity.
When presented with the logical outcomes of viewing and treating biological species as Class constructs
in our current Darwinian worldview of descent with
modi¢cation that has structured so much of scienti¢c
thought, it is untenable to accept any of the standard
concepts of species save the unique nonoperational
ESC.
The dichotomy of these two approaches to viewing
and treating species is identi¢ed in Fig. 1(B) wherein
species as Classes have no temporal component,
do not change, and lack any (including historical)
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
181
On species and species concepts R L Mayden
Figure 2 A series of phases of an investigation of biological diversity (I^IV) and comparison of the radically di¡erent
conclusions when one chooses to view species as Class constructs (Phases I, II) or Individuals (Phases I, III, IV). In this example
the critical di¡erence is the approach chosen by an investigator to investigate the origin or an intermediate genotype in the
middle two metapopulations of a new species. If one views species as simply Class constructs then comparisons of this new
species with northern and southern species results in a logical interpretation of the geographically intermediate
metapopulations being intergrades. However, if one views species as Individuals and lineages then the origin of the
intermediate genotype, when viewed from a phylogenetic perspective, can be interpreted as simply the retention of as
plesiomorphic genetic variability ^ requiring no scenarios about a fabricated ongoing process that ultimately leads to our loss
of biodiversity.
cohesion. However, species as Individuals possess
these attributes. In a ¢nal example illustrating the
gross di¡erences between these two philosophies,
we will look at a heuristic example of diversity. A
researcher discovers two populations or metapopulations of a taxon that she/he believes represents a
unique evolutionary lineage and describes it as a
new species (Fig. 2, Phase I). Later, the same researcher examines genetic variation in this new species and discovers that it possesses alleles A and B in
both metapopulations (Fig. 2, Phase I). A second
researcher examines closely related species in adjacent northern and southern regions of the new species and ¢nds that the southern species possesses
182
onlyallele B and the northern species possesses allele
A (Fig. 2, Phase 2). At this time, based on no phylogenetic or historical perspective to interpret these patterns of character variation, the second researcher
argues for large-scale introgression between the
northern and southern species and that the new
taxon described by the ¢rst researcher is nothing
more than populations of intergrades, thus falsifying
the hypothesis of a unique independent lineage
(Fig. 2, Phase 2).Without a historical or phylogenetic
perspective to interpret the origins of the character
data the second researcher, without possibly knowing, has just treated these populations/species as
Class constructs, not Individuals. Interestingly, this
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
Nature produces Individuals that participate in processes. We use these Individuals to pose questions,
hypotheses and theories, from which we make observations, generate new hypotheses, and refute or
corroborate existing hypotheses. The intermediate
between questions and investigation phase and the
phase wherein we develop conclusions, test theories
and hypotheses, and generate new hypotheses is the
biological species. Thus, how we Individuate things
in our natural world will have major implications on
our science! In this sense, the quote by E.O. Wilson
(1992) in The Diversity of Life is, particularly apropos:
‘the species concept is crucial to the study of biodiversity . . . Not to have a natural unit such as the species
would be to abandon a large part of biology into free
fall, all the way from the ecosystem down to the
organism’. The need to emphasize that species as
Individuals are lineages and can be identi¢ed using a
variety of data types and analyses and falsi¢ed as
scienti¢c hypotheses cannot be emphasized enough.
This basic understanding is lost, however, in much
of the discussion of species concepts and is also
ignored by many having tired of the controversy.
Rather, they prefer to recognize taxa on the basis of
what they think is best. In many cases, this method
probably serves well in identifying biodiversity for
the experts on particular taxonomic groups. However, as evolution proceeds at di¡erent rates and on
di¡erent scales and characters, others may have important information to re¢ne our ideas of diversity.
The bottom line is that viewing and treating species in nature as Classes destroys one’s ability to work
harmoniously within the theory of descent with
modi¢cation. It prohibits one from understanding
the lineage, temporal, and spatial explanations for
the origins of characters or attributes possessed by
species. Under this paradigm many errors accounting for biological diversity will occur. In fact, in the
recent book Biodiversity: A biology of numbers and distance (1996) chapters by Gaston (1996a, b) and Mallet
(1996) both miss the signi¢cance of species as Individuals and damn the controversy over species by
essentially calling for strictly phenetic methods by
completely untrained personnel to delineate diversity. Both of these authors also miss the point of species as lineages delineated by multiple types of
characters under di¡ering rates of evolutionary
change. Such attitudes of species as Natural Kinds
will most emphatically have a severe negative e¡ect
on any scienti¢c discipline related to biodiversity.
Finally, treating species as Classes also has longterm e¡ects on the perception of biological diversity
in natural systems. Naturally, most ecologists, behaviourists, ¢sheries biologists, physiologists, etc. are
not particularly skilled at the nature of species or
even biological diversity. Their knowledge of the nature of species may be basically limited to the Linnaean category species, which is a Natural Kind or a
Class construct into which we place species in nature
thought to participate in natural laws governing
such entities. This limited perspective can have a
major e¡ect on how science is done in these disciplines. As an example, in many ecological studies,
species and organisms are typically grouped into
functional groups like scrapers, shredders, ¢lter
feeders, etc. as examples of feeding groups. As such
they create groupings of Individuals into Class constructs and discuss the generalities of the Classes in
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
183
standard Class-like interpretation of character variation or character validity does not view species as
entities possessing traits because of descent from
common ancestors participating in natural selective
processes wherein features may be gained, lost, or
modi¢ed through time.When viewed from this extremely limited historical perspective the levels of natural biological diversity will both be less than what
actually exists and will decline with increasing revisionary studies conducted outside of a phylogenetic
framework. Alternatively, the ¢rst researcher examines the character variation in a phylogenetic context
(Fig. 2, Phases II and III). Through this process, the
researcher discovers that the occurrence of both
alleles A and B in the centrally located new species
owe their origins not to active or historical introgression, but to the retention of the plesiomorphic allele
B in the new species and the southern species and
the origin of a new allele A in the common ancestor
to the new species and the northern species. The lack
of allele B in the northern species is simply a matter
of ¢xation of allele A in this population. This interpretation is only possible because it is rooted in a
researchers perception of species and lineages as
Individuals evolving through time and participating
in natural processes of descent, including speciation.
Thus, practitioners of biological diversity, at any level,
must have an appreciation for and understanding of
the knowledge of phylogenetic systematics and species (and other elements of diversity below species)
being Individuals. Otherwise, inappropriate decisions
by untrained‘specialists’will occur.
Why then are Individuals so important to
clearly identify?
On species and species concepts R L Mayden
ecosystem dynamics, food webs, energy £ow, etc. In
order to be a member of one of these particular
Classes, each species must have the required attributes set forth in the de¢nition. The general conclusion to many of these types of studies has been that it
does not really matter which particular species of
the Class ‘¢lter feeder’ or ‘predator’ are present in a
system, there simply need to be ¢lter feeders, predators, scrapers, or whatever in communities. This type
of reasoning is extremely damaging to ecosystems
and communities because the investigator is not cognizant of the inherent uniqueness of di¡erent
lineages that evolved at di¡erent times and places, as
would be implicit with the education of species as
lineages. Treating species as Classes that ¢t into ecological or other guild Classes completely removes
any temporal and spatial, as well as historical, components from these interpretations. Sadly, such logic
may eventually determine or be used to argue that
only a few species worldwide are really needed to
serve as functional communities!
Thus, although those continuing to acknowledge
the dialogue of species as Individuals may be viewed
as eccentric, and the philosophical dialogue as inapplicable to science, they lack an appreciation for how
the di¡erences between Individuals and Classes will
have signi¢cant underlying implications to inquiries
in science and other disciplines. In my opinion, once
armed with this knowledge, advancing the many
¢elds of the sciences will come naturally through
empirical studies, analyses and interpretations.
The hierarchy of concepts of species
A hierarchical view of concepts of species provides a
functional and productive means of avoiding the
paradoxical notion of treating species as Classes
while aspiring to discover and understand natural
processes involved in evolutionary history, patterns
of speciation, or the tree of life. As previously proposed (Mayden 1997, 1999), the only sound method
to account for the diversity of life is through the guidance of the ESC as a primary concept. As the only
concept not precluding the existence of some independent lineages thought to be species because of
operational, pragmatic, value-driven, or aesthetic
concerns for particular types of traits or divergence
levels in these traits, it is the only concept currently
available to serve as an over-arching idea as to the
notion of biological species. This concept is free of
the baggage encountered in all other concepts requiring particular kinds or qualities of essence being
184
present. Hence, Individual-like entities are acceptable
or permissible under this concept as biological species regardless of the types of changes in characters
or homologues or inferred processes (mate recognition system, reproductively isolated) that have
occurred to substantiate the central hypothesis of
lineage independence.
The only problem with employing a nonoperational concept is that it is di⁄cult for anyone to ¢nd
anything to ¢t the concept without prior knowledge or guidelines or tools to use in the discovery
phase. As one solution to this seemingly intractable
dilemma of trying to discover entities that lack de¢nition, I proposed that most other hypothesized concepts of species are consistent with the primary
concept and outline minimal standards for species
recognition, and that these be considered as surrogate concepts to the ESC. Using these concepts as
our ‘toolbox to biodiversity’, researchers can use
these various concepts, where appropriate, to discover lineages consistent with the primary ESC. In
this framework, researchers using any of the secondary concepts can identify species diversity based on
criteria of homologueous divergence in morphology,
ecology, behaviour, genetics, or any other quality
supporting the hypothesis of lineage independence.This
argument was adopted and supported by Stau¡er
and McKaye (2001).
Advocating this position, I acknowledge that I am a
monist with respect to a concept of species, that is
the ESC. However, I am a pluralist with respect to
advocating multiple concepts of how to identify entities in nature that are consistent with my preference
for the primary species concept. To some, my position
may seem reasonable, but to others, maintaining
both a monistic and pluralistic view may seem unacceptable. However, let us examine a simple real-world
example illustrating the continuity and harmony of
my position with the species concept issue. In this
case, it involves the widely accepted concept of monophyly and surrogate concepts used to aid in identifying monophyletic groups.
A parallelism of monophyly and the
evolutionary species concept
Throughout modern systematic literature, the term
monophyly has signi¢cance to not only systematists
but to many other scientists from varied disciplines,
probably second only to the term synapomorphy.
Today, the recognition of supraspeci¢c taxa requires
that the grouping be monophyletic, that is the group
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
must include the unique common ancestor and all of
its descendants. Other groups, those either not
including a unique common ancestor or excluding
some of the descendants of a unique ancestor, are
not recognized because they are arti¢cial constructs
that are completely misleading to anyone looking to
biological classi¢cations for insight into evolutionary
history of the tree of life. Interestingly, as discussed
above, species are the initiators of all supraspeci¢c
monophyletic groups and all monophyletic supraspeci¢c taxa are special types of Individuals (Wiley
1981). Likewise, all nonmonophyletic taxa (para and
poly-phyletic groups) are arti¢cial constructs and
can be referred to as arti¢cial Classes, but not Natural
Kinds. Although monophyletic groups are often
referred to as clades, these latter groups are often
referred to as grades or grade groups (arti¢cial constructs).
Interestingly, although the term and concept of
monophyly receives worldwide acceptance as an
appropriate concept for supraspeci¢c taxa, there is
no one thing operational or pragmatic about this de¢nition. Given the de¢nition of monophyly as a
describing a theoretical entity (much like biological
species), one would be hard pressed to go out, use the
de¢nition, and return with a monophyletic group.
Why? Because it is a theoretical concept characterizing a special type of by-product of nature, a special
type of Individual referred to as a Historical Group
(Wiley 1981). It is a type of Individual that originates
as a biological species and participates in natural processes that result in additional Individuals, none of
which can be de¢ned in the traditional sense, only
diagnosed.What are the diagnostic traits of a monophyletic group? They are the same traits that were
diagnostic of the unique common ancestral species
(Individual) to the monophyletic group. Lacking an
ability to discover monophyletic groups, scientists
must bridge over to surrogate concepts that are consistent with the over-arching concept of monophyly,
just as with the ESC and the alternative pragmatic,
operational concepts of species.
But, as with the conundrum of biological species,
what operational, purely pragmatic concept do we
choose to identify monophyletic groups? Theoretically, there are many ways to discover monophyletic
groups, or groups of species thought to be derived
from a unique common ancestor. One may argue that
overall similarity is a good criterion, that genetic or
morphological dissimilarity and discontinuity are
good criteria, that taxa engaging in sexual reproduction following distinct mating rituals of some sort
can distinguish groups, or that only things with distinct colour groupings can form monophyletic
groups. We may use any or all of these criteria for
monophyly and will ¢nd that some diversity may be
excluded by each of these concepts or that these concepts lack logically sound justi¢cations for identifying monophyletic groups. In fact, in the history of
systematics di¡erent criteria have been used to justify
groups recognized in the Linnaean Hierarchy (even
phenetic similarity) and today the concept most often
employed is that of synapomorphy, that is, monophyletic groups are supraspeci¢c taxa containing Individuals thought of as species that can be identi¢ed and
grouped on the basis of at least one shared^derived
character that they inherited via their unique common ancestor. This is a character that evolved in the
common ancestor to the group and passed on to the
descendants of the group (or parts of the whole) as
homologueous characters. The most commonly
employed scienti¢c method for identifying synapomorphies is the process of outgroup comparison
(Wiley 1981). Thus, systematic biology searched and
tried various ways to logically discover meaningful
methods to identify monophyletic groups consistent
with the theory of descent with modi¢cation. We do
not want to preclude groupings because of bias
against particular types of characters or degrees of
di¡erentiation as long as groups being recognized
are consistent with the idea that natural lineages
and groups of lineages exist as a historical process.
The recognition of this natural order versus arti¢cial
constructs developed not from scienti¢c principles
or processes but by simple value judgements and aesthetic desires, is bene¢cial to the scienti¢c community. Today, the synapomorphy concept is most
widely accepted as a criterion for recognition. This is
not to say that other criteria may not work or will
not be discovered in the future. In fact, if one knows
that all traits being examined for a particular group
of organisms evolved at a constant rate (anagenesis),
or if one knows the particular model of character
evolution in the evolutionary history of a group, then
overall phenetic/genetic similarity or probabilities
assigned to nodes in likelihood analyses, respectively,
would work equally well for identifying monophyletic or natural groups. Unfortunately, assumptions
on rate changes of characters and models of character evolution are dangerous because they are not
known or predictable.
Thus, it is clear from this parallel situation that the
biological community identi¢es the signi¢cance of a
hierarchical structuring to identifying monophyletic
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
185
On species and species concepts R L Mayden
groups wherein criteria like synapomorphy serve as
acceptable surrogate means for discovering clades.
In the same vein, the hierarchical arrangement of
concepts of species identi¢es the ESC as the primary
concept and almost all other concepts as operational
surrogates for discovering and hypothesizing the
existence of species in nature. Some concepts, however, are not acceptable for identifying species (Successional Species Concept) and some others should
be used with caution to avoid recognizing arti¢cial
constructs (Mayden 1997, 1999). In my opinion, the
most useful surrogate concept for the ESC is the PSC
(all versions, Table 1), although caution must be
employed even with this concept. It is clear that none
of the PSC concepts will allow one to discover ancestral species and if one employs only the diagnosable
version one may recognize arti¢cial groupings.
Data chauvinism, aesthetics and evidence for
the existence of a species
Following from and directly related to theoretical and
philosophical issues of species (or monophyletic
groups) is the question of what constitutes the ‘best’
methodologies and empirical data for identifying
species? This particular topic is one that is rarely discussed within the realm of species concepts because
of the inherent nature of species being Individuals
and their lacking de¢nitions. Given that species
descriptions are hypotheses presented to the scienti¢c community regarding a unique and independent
lineage in our biodiversity what type of data and
methodology does one need to propose such hypotheses? Species concepts themselves are hypotheses
regarding the nature of species as being reproductively isolated, morphologically distinct, genetically
distinct, having phenetic similarity, etc. Basically,
what is it that a practicing scientist in systematics
and taxonomy charged with discovering and documenting the diversity of life looking for? In my opinion we should be looking for Individuals consistent
with the ¢rst order principles of the Darwinian paradigm of descent with modi¢cation wherein tokogenetic relationships exist within a lineage and the
lineage maintains cohesion and independence from
other such lineages ^ evolutionary species. As such,
these lineages are subject to selection, anagenesis
and speciation and many other evolutionary
mechanisms that operate within lineages but not
between them.
What then is subject to falsi¢cation in this scienti¢c process? This depends upon the frame of reference.
186
Clearly, the hypothesis of a particular species being
described by a researcher is a testable hypothesis that
can be falsi¢ed. However, the various species concepts themselves should be and can be the subject of
testing and falsi¢cation. For example, systematists
traditionally reject the hypothesis of the chronological or successional species concepts as valid hypotheses for recognition of independent lineages.
This concept, while functional for biostratigraphers,
knowingly recognizes di¡erent ‘species’ transforming within a single lineage.Thus, other concepts used
to recognize valid species as independent lineages
participating in evolutionary processes should also
be subject to falsi¢cation. As discussed in earlier
papers (Mayden and Wood 1995; Mayden 1997, 1999;
Wiley and Mayden 2000a,b,c) the great diversity of
species recognized today should be used to test the
validity of the di¡erent concepts as to their ability to
account for the diversity of independent lineages.
All of the secondary concepts would fail such a test
because of their emphasis on particular pragmatic
necessities for recognition above and beyond the
basic premise of an independent lineage. Some species diversity, hence biodiversity, will necessarily be
excluded from recognition as valid evolutionary entities that should be recognized for various reasons.
What about hypothesized species? How does one
test the validity of a species as an independent lineage? Simply, evaluate the characters being used to
denote it as an independent lineage, as well as other
traits appropriate for testing lineage independence,
within the proper methodological framework. If one
cannot falsify the hypothesis presented by the original author(s) that the biological entity represents an
independent lineage, then one cannot reject the species status. Further study, however, may provide
additional evidence for lineage independence that
would corroborate the original hypothesis (but, see
caveats below).
Does a particular concept of species permit one to
falsify the hypothesis of a distinct species formulated
under a di¡erent hypothesized conceptualization of
species? Emphatically, no, hypotheses of species
stand alone independent of particular hypotheses
outlined as species concepts. It has been clearly
demonstrated that except for the ESC, all current species concepts may be rejected as sole hypotheses of
species in nature if subjected to falsi¢cation by all
currently accepted species described from nature. A
researcher employing one concept of species will propose new species thought to represent independent
lineages based on that concept. Other researchers
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
identify independent lineages based on other concepts.
If there is con£ict between these hypotheses it will
likely be because of the a priori requisites of the species concepts and not because of inadequate evidence for proposing lineage independence. It is here
that the area of species concepts and species recognition departs from science and moves into the nonscienti¢c realm involving value judgements and
aesthetic assessments. Using the hierarchical method
of primary and secondary concepts, however, species
recognized under one concept are just as valid as
those recognized under other concepts, as long as
they are consistent with the ESC. Under this model,
species as lineages are the hypotheses to be tested,
regardless of the concept that one employs.
Descriptions of species should be well-formulated
hypotheses of independent lineages that are hypothesized to exist into the future until such time that
they become extinct. No particular person has a crystal ball that can predict the future of an entity that is
clearly an independent lineage today. It may be that
such a lineage may merge with another lineage
through introgressive hybridization in the distant
future. It may also maintain its independence even
in the face of di¡ering levels of hybridization. Thus, it
is impossible for a researcher, armed with multiple
technologies, to decide between these situations and
value assessments as to whether one lineage is a species or that another is not is inappropriate in the
scienti¢c realm.
What kinds of evidence should be used to individuate species as lineages? It is also here that values and
aesthetics are inadvertently involved in proposing
and testing scienti¢c hypotheses. All types of heritable characters capable of identifying distinct, evolutionarily independent lineages are appropriate for
hypothesis generation and testing. Obviously, this
has not always been accepted and will not likely be
completely accepted in the future. Why? Because as
humans, we have a tendency toward assigning personal values or aesthetic priorities. Characters used
to identify lineages and their use in determining
appropriate level of recognition of naturally occurring entities are neither immune to our imposition
of these priorities that have little to do with the discovery process of science. Preference is given towards
characters that recognize species on the basis of morphological divergence, traits that are more easily
observable and leave little doubt in the minds of those
capable of seeing these characters. Behavioural characters have traditionally been taboo in systematics
and taxonomy for recognition of both biodiversity
and their phylogenetic relationships (Stau¡er et al.
2002). Why? Because, for many years, dogma held
that these character types were very labile and not
reliable. Interestingly, the advent of many new methods for either enhancing our observational skills
(microscopes vs. hand lenses) or providing new information pertinent to the lineage independence question (proteins, DNA) has led to mixed methods for
diversity questions. Unquestionably, cryptic biodiversity does exist, as species that look essentially identical for us but maintain independent lineages.
However, alternative attributes and techniques (behaviours, mating systems, ultrastructure, pheromones, molecular techniques, etc.) are capable of
sorting out independent lineages identi¢able as species. Oddly enough, some of these lineages enjoy
much attention in the scienti¢c arena because of the
evolutionary peculiarity that they appear identical
to us but are not the same to each other! In other
cases, however, independent lineages are identi¢ed
based on alternative traits to morphology but are not
recognized as species, largely because of the dominance of morphological attributes for diagnosing or
keying out species. Many recent molecular studies
have revealed many new evolutionary lineages that
are consistent with the primary objective of discovering species diversity. However, few of these are ever
recognized, an exception being the work by Highton
et al. (1989) molecular study of salamanders of eastern North America.The inconsistency in recognition
of biological diversity clearly identi¢able as independent lineages and equally justi¢able as those based
on morphological traits lacks scienti¢c rigor and
derives from values and aesthetic assessments imposed by each scientist.
Rank order in taxonomic systems, based on degree
of di¡erentiation, while largely eliminated from systematic biology in the recognition of supraspeci¢c
taxa is commonly employed in studies investigating
species diversity. As with the inappropriate use of
value and aesthetic assessments described above,
researchers frequently argue over the validity of species based on the degree of di¡erentiation that a species may possess for a trait or the type of trait being
used in recovering diversity. While these arguments
do not pertain to evidence for lineage independence
and do not falsify initial hypotheses of species, these
are used to determine whether lineages should be
considered variants, races, subspecies, evolutionarily
signi¢cant units (ESUs) or species. Essentially, the
more divergent a lineage is, mostly based on morphological evidence, the higher the ranking. Why?
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
187
On species and species concepts R L Mayden
Because researchers feel more comfortable in recognizing lineages that are more divergent and are lead
into a false sense of security and comfort that these
lineages are more likely to survive the test of time.
These lineages are assigned higher value in evolutionary biology and general public opinion because
they are given proper nouns, receive recognition as
elements of selection and speciation, and even have
higher priority in listing and protection when endangered. Lineages recognized solely on the basis of
molecular, biochemical, physiological, or behavioural
traits have little or no chance of being recognized in
the current philosophy because researchers do not
¢nd these types of lineages as aesthetically pleasing
to their sense of being or existing in nature as real.
These same lineages, if recognized, are usually not
named or are given lesser rank than are taxa that are
recognisable on the basis of other types of data, typically morphological. However, evenwithinmorphological analyses lineages identi¢able on the basis of
some types of characters are ultimately deemed less
valuable or less aesthetically pleasing than those
based on other types of characters. In ¢shes, lineages
that di¡er only in complete di¡erences in the number
of anal ¢n rays counts are less likely to be recognized
than those di¡ering in colour patterns and anal ¢n
ray counts. Both of these sets of evidence may provide
exactly the same information supporting lineage
independence; the only di¡erence is an arbitrary
assessment of value and aesthetic palatability by the
scientist or the panel of experts evaluating what constitutes recognisable diversity.
Are some types of data better in the scienti¢c expedition of discovering diversity? No, as long as the
evidence is heritable and provides biologically
meaningful information for lineage independence.
As discussed above, the disciplines of taxonomy and
systematics have a history of favouring morphological types of characters and certain degrees of di¡erentiation. One may argue that this is inherently
linked to two things, use in identi¢cation and the
need for diagnoses in species descriptions. True, morphological di¡erentiation at signi¢cant scales does
make it easier to identify diversity. It is a more convenient tool for identifying species given the inherent
limitations of the senses of perception of humans.
However, is this the operational criterion, the value
component, or the aesthetic assessment that we
should be using in delineating species as elements
of our biodiversity. I would think not. Again, evidence for lineage independence, cohesion and continuity derived via whatever biologically meaningful
188
methods or tools should all be considered on a equal
playing ¢eld. There are at least four caveats to this
statement. First, any comparisons or tests of hypotheses of lineage independence of species must be cast
in a phylogenetic perspective wherein researchers
are cognizant of the existence of both plesiomorphic
and apomorphic traits existing in lineages. Thus, the
existence of plesiomorphic traits in a lineage is not
necessarily indicative of a lineage not maintaining
its independence from close or distant relatives. Second, any comparisons of species validity must be
between sister species derived from a unique commonancestor. Comparisons between distantly related
species regarding lineage independence is meaningless given that lineage independence results from
speciation events producing sister species. Third,
comparisons must be cognizant of known di¡ering
rates of evolution or anagenesis of within and
between di¡erent types of characters that occur
both within and between lineages. Fourth, species
as Individuals can maintain their lineage independence, cohesion and continuity even in the face of
observed cases of sexual transgressions across species ‘boundaries’ (Wiley 1981; Dowling et al. 1989;
Warren1992).
Regarding character superiorityand rates of evolution, today many biologists and laypersons alike tend
to view molecular studies and data as having ‘more
strength’ than other types of studies involving behaviour, physiology, ecology, and even morphology in
‘testing’ species hypotheses. This prejudice relates
directly to an ignorance of the inherent and routine
problems one encounters with generating and interpreting molecular data, a problem often not realised
in such studies or the intended audiences. For example, it is commonly believed that when two entities
share the same feature they are said to be the same;
two species with the same mobility of an allele at a
particular locus are thought to be the same, or the
occurrence of the identical allele provides ‘evidence’
of sameness of the two or more taxa. This mode of
interpretation permeates all types of data but is especially prominent in studies involving molecular data.
Little thought is given to the well-corroborated evidence that di¡erent types of traits (and genes being
sequenced) evolve at di¡erent rates over time, and
that even within the evolution of a gene, the rate of
divergence can vary. This is shown in Fig. 3 where
genes A and B exemplify the common view of gene
evolution by individuals holding that similarity
equals sameness. Both of these genes have a constant
rate of change over time, but gene A is evolving at a
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
Figure 3 Rate of change for six hypothetical genes that
can be used to delineate species. Genes A and B are evolving
at a constant rate while genes C^F evolve at di¡erent rates
depending upon the evolutionary time period examined.
rate higher than gene B. However, in this example
genes C^F have di¡erential rates of evolution depending upon the time period when examined.Within an
evolutionary perspective this means that one cannot
rely upon a general ‘molecular clock’ to predict or
de¢ne the ‘amount of divergence necessary’ before a
biological entity is considered a species! Thus, it is
natural that di¡erent species will have di¡ering levels
of genetic di¡erentiation (¢xed genetic di¡erences,
di¡erences in morphological divergence, di¡ering
genetic di¡erences). There is no magic amount of
divergence that should be used to declare species
validity!
This view of character divergence and rate of evolution is also illustrated in Fig. 4 where hypothetical
speciation events are identi¢ed along the time scale.
Here, the arrows indicate the timing of speciation
events corresponding to the phylogenetic hypothesis.
Each of these events also have particular rates of
change in genes A^D associated with them before, at
and after each speciation event. In this example, only
genes C and D would meet the general assumptions
of a molecular clock or a generalized criterion for
genetic divergence necessary for speciation. For
these two genes, the untrained biologist is faced with
one gene providing more evidence of divergence
(gene C) than the gene evolving at a slower rate (gene
D). The other two genes are evolving at variable
rates over time. The pattern of evolution in all of
these genes o¡ers explanatory evidence as to why
in some groups of organisms a particular gene
evolves at a slower or faster rate than in other groups.
Thus, a strict reliance upon particular levels of divergence in particular types of characters, whether
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
Figure 4 Di¡erential rates of change in four hypothetical
genes used to delineate species and the points in time where
speciation occurred. Of these, only genes C and D meet
traditional expectations of constant rates of change and the
perception of a‘molecular clock.’ Even these genes would
lead some biologists to di¡erent conclusions. Genes A and B
are variable in their rates of change, providing evidence
against a‘molecular clock’and the use of absolute
di¡erences for species delineation.
morphological or molecular, is very problematic in
identifying biological diversity.
Implications of species concepts on the
diversity of fishes
Quite clearly, the above essay on species as Individuals, the hierarchy of concepts under the ESC, and
the logical nonbiased interpretation of all types of
data to identify and test for lineage independence,
cohesion and continuity will have major, long-lasting
implications to the diversity of ¢shes and other
groups. The arguments for this ideology are already
well developed but have not been implemented. There
is some hesitance in the community to employ nontraditional tools to validate the existence of independent lineages of diversity. To some degree, this is
logical in the sense that describing species on the
basis of molecular (or other) characters alone may
preclude the identi¢cation of species using traditional characters. This is a practical concern that I am
con¢dent many have considered. However, we must
also consider the impact of not recognizing any type
of diversity that clearly is consistent with the logic
for describing species based on acceptable morphological characters. That is, if our objectives are to discover natural patterns of biological diversity and use
this information to understand natural processes
responsible for its evolution, what good is it for us to
only recognize some types of diversity, but not others,
189
On species and species concepts R L Mayden
when they all can be justi¢ed equally well? Interpretations of pattern and process will necessarily be
fatally £awed. Our understanding of these processes
and resulting patterns will be completely contingent
upon priors.
The impact of not recognizing independent lineages of diversity is probably most noticeable in conservation biology with the recent increase in the
destruction of ecosystems and extinction of species.
Recognizing species as Individuals and thinking of
them as lineages related to other lineages is critical
to conservation e¡orts (Mayden and Wood 1995; Roe
and Lydeard 1998). Conservation strategies for the
recovery of species must be soundly based and, in
my opinion, must incorporate phylogenetic information as to the sister-group relationships of species
(Mayden andWood1995). Using sister-group relationships, one can identify sister species that can be used
as surrogates for extensive investigations inquiring
as to the nature of the imperilment of an endangered
species.When species are endangered it is customary
to do certain things. These include: (i) protection of
the species, but not necessarily its habitat or ecosystem, so that no one can sample or harass the species;
(ii) a type of ongoing status survey of the species; and
(iii) and inaction to see what happens, all the while
hoping that the species status will improve. This is a
very passive approach to a problem of unprecedented
proportions relative to ecosystems and species diversity. Alternatively, if one views the problem with species being Individuals behaving as lineages and
sharing common ancestors with other closely related
species, then we may use this information in the
development of a more proactive and pro¢table conservation strategy. Sister species can and will be used
as surrogate species for detailed investigations and
through phylogenetic methods involving characterstate reconstruction, we may predict with a high level
of con¢dence what the underlying ecological, physiological, behavioural, etc. ‘weaknesses’ are for an
endangered species in a particular habitat.This is conservation through phylogenetic surrogacy. Agencies
and researchers need to acknowledge natural diversity recognisable as independent lineages on the basis
of multiple types of data.They must acknowledge and
incorporate a lineage and phylogenetic perspective in
their recovery plans to better serve their objectives.
A glass ceiling on species diversity
Undoubtedly, many reading this essay will cringe
from the thought of recognizing all independent
190
lineages as biological species.Who knows how much
diversity this may result in, as we perfect our methods of better understanding biological diversity. However, under the ESC as outlined most recently by
Wiley and Mayden (2000a) species are characterized
as follows: an evolutionary species is an entity composed of organisms that maintains its identity from other
such entities through time and over space and that has
its own independent evolutionary fate and historical tendencies.This concept is further ampli¢ed in this paper
to cover all types of questions regarding varied kinds
of entities and how they are interpreted. But, one
thing must be clearly understood. This is a lineage
concept of ancestor^descendant populations, metapopulations, or other groupings; it is not a concept
that should be abused by those that may critique the
lineage idea wherein ‘anything’ could be diagnosed
as a lineage using the right techniques. Such arguments are biologically meaningless and only cloud
the important issue of striving to recover natural patterns of descent.
Yes, it is very likely that the number of species that
will be recognized by employing the ESC will increase
relative to the numbers of taxa that have been recognized using other concepts, especially the BSC. In a
real example resulting from phylogenetic studies
and species diversity, Cracraft (1992) found that the
number of species of birds-of-paradise increases from
40 under the BSC to 90 using the best surrogate concept for the ESC, the PSC.Why is this? Because other
concepts severely restrict the types of biological
diversity that is allowed to be recognized. Ironically,
probably the most widely know concept, the BSC,
represents one of the biggest impediments to not only
recognizing diversity but also conserving diversity.
This concept, often referred to as the polytypic species concept, resulted in a tremendous ampli¢cation
of taxa referred to as races or subspecies in the
1960s. As members of the same species, populations
of these ‘polytypic species’ in allopatry were often
transported and mixed with other populations of
the same or di¡erent subspecies. This is often a problem in ¢shes, particularly those of importance to
¢sheries and anglers. The result of many of these
e¡orts to maintain populations in di¡erent areas by
mixing di¡erent ‘subspecies’ has resulted in the loss
of what are now known as divergent lineages, especially in western North American trout (Miller et al.
1989; Behnke 1992). In fact, some ‘subspecies’ or
‘races’of trout were actually found in the same lakes
and in sympatry, and some have gone extinct as a
result of introgressive hybridization with introduced
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
taxa. The implication that, of all the independent
descendent lineages that exist in nature, only
some can be recognized as species, justi¢ed on the
basis of value judgements, aesthetic priorities, or
even con¢dence of eventual lineage mixing of two
‘subspecies’ in ‘potential contact’, is not in the realm
of science. This logic can and will have detrimental e¡ects on many, many disciplines working with
biological diversity, not to mention the diversity
itself.
Today there is a glass ceiling on species diversity
driven largely by nonscienti¢c values, aesthetic
assessments and preferences, and operational issues.
While predictions as to the number of species di¡er
widely between authors, ultimately the recognition
of diversity rests in the information that is permitted
to be published. If we, as practising scientists, knowingly exclude biological diversity, proposed and published to represent valid evolutionary species, on
the basis of unpublished opinions that may or not
have any scienti¢c rigor and validity, we depart from
science. Rejection of species hypotheses must be
based on well-formulated studies that demonstrate
falsi¢cation of the lineage hypothesis and must be
unbiased with respect to the nature of the characters (as long as the characters are heritable and
homology can be ascertained) and free of a propensity of value and aesthetic bias. Historically, some
publications, books and checklists driven by committees or panels knowingly exclude hypothesized
species diversity, or decide what are the valid species
and which ones should be relegated to ‘subspecies’
or not recognized at all (Robins et al. 1980, 1991).
Although some may argue that this is done with
good intentions for accounting for only the ‘good’ or
‘valid’ species or to ‘simplify’ complex systems, such
decisions should not be made without presenting
adequate data to the scienti¢c community to
demonstrate alternative views. At least for North
American ¢shes this trend is beginning to change
with an improved understanding and appreciation
for the diversity of species concepts. This has
recently been demonstrated by Warren et al. (2000)
wherein criteria for species inclusion were clearly
outlined.
Today we live in a world of incredible technological abilities and the number of species recognized
should not be constrained to those that can only ¢t
in an atlas or those that are morphologically divergent for preferred character types within particular
higher taxa. Experts of di¡erent taxonomic groups
ultimately determine the diversity of species within
any one group, but this should be done with peer
review. As experts, we provide fundamental knowledge that other disciplines rely on for their own
investigations into a vast array of data- and theorydriven inquiries of the physical and natural worlds.
We hold the responsibility of providing sound scienti¢cally based estimates of biological diversity referable to species and we must make this information
readily available to the world.We must aspire to provide these unbiased estimates of evolutionary
lineages regardless of di¡erences of opinion regarding nonscienti¢c issues, and we must successfully
achieve these goals that our disciplines have pledged
to the scienti¢c and nonscienti¢c communities. In
doing so, systematics and taxonomy as scienti¢c
endeavors will continue to maintain its fundamental signi¢cance and its deserved respect in these
communities.
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
191
Returning to the hierarchy of concepts
As outlined above and in previous papers (Mayden
1997, 1999) the Evolutionary Species Concept serves
as a logical and fundamental over-arching conceptualization of what scientists hope to discover in nature behaving as species. As such, this concept with
its lack of operational necessities can be argued to
serve as a primary concept of diversity. Also as discussed, however, it is impossible to use this concept
to go out and operationally discover or recover entities consistent with the ESC. As in the analogous
situation between monophyly and synapomorphy
described above, we must implement means by
which species can be discovered, described, and diagnosed as hypotheses of lineage independence, cohesion and continuity subject to falsi¢cation. Clearly,
the synapomorphy serves as a functional guide or
part of our phylogenetic ‘tool box’ best suited to identify monophyletic groups, given our current understanding of phylogeny reconstruction. From this, it
logically follows that the numerous secondary species concepts in actuality do not have the same ontological status as the ESC and are subservient to it,
serving as functional guides or part of our lineage or
species ‘tool box’ best suited to identify natural
lineages thought to represent species, given our current understanding of the patterns and processes of
descent with modi¢cation. Thus, in reference to this
fundamental di¡erence between the ESC and secondary concepts, I do not consider them to represent
fundamental concepts but simply guidelines. As
such, I recommend that these alternative concepts
On species and species concepts R L Mayden
be referred to as methods, guidelines, or prescriptions
for the di¡erent types of diversity that our current
and ancestral academic ancestors have developed
for detecting species, derived out of di¡erences in
taxonomic groups, and preferences for di¡erent types
of criteria. None of these guidelines, however, should
be acknowledged as scienti¢c if they knowingly permit the recognition of false taxa (Successional Species Concept) or if they or their applications are
laden with nonscienti¢c value- or aesthetic-based
arguments (Table 1).
Individuals above and below the scale
of species
The idea of Individuality has focused largely of late on
the species issue and has brought considerable
debate to the topic that has lead not only to multiple
conceptualizations but also questions as to the scienti¢c rigor of the scienti¢c discovery process in systematics and taxonomy with species. While only
brie£y addressed herein, the issue of Individuality,
Individualism and Individuals is not restricted to the
species. In fact, entities ¢tting the characterization
of the philosophical term Individual dominate many
other disciplines of sciences. Characteristically, one
can easily identify such issues, among many other
means, by looking for those things in disciplines that
have an inordinate number of de¢nitions and are
controversial. The fact that entities thought to be
naturally occurring by-products of some natural process have multiple de¢nitions is a clear indication of
the fact that the boundaries of these entities are not
clear but are ‘fuzzy’ (Fig. 1). Why are they fuzzy? For
the same reasons that species boundaries are fuzzy,
they change over time, they do participate in natural
processes governing these bodies and, probably most
importantly, as Individuals they cannot be de¢ned ^
they can only be described or diagnosed. Where are
these nonspecies examples? Only to list a few in an
order of scale, things such as galaxies, stars, planets,
biomes, ecosystems, communities, wetlands, populations, organisms, organs, organelles, molecules and
atoms are all Classes that contain Individuals and will
necessarily invite controversy when boundaries are
to be made (similar to the taxonomic category Species as a Class containing Individuals known as species). Taking the planet Earth as an example, where
is the boundary of Earth such that investigators
wishing to better understand processes can limit
their collection of data? Is it the geological surface,
the geological and aquatic surfaces, the edge of the
192
stratosphere or beyond? For ecosystems, communities and, in particular, wetlands where does one
begin and end? Depending upon the multiple de¢nitions of a wetland the boundaries can vary by several
meters. Given that there are multiple de¢nitions as
to the boundaries of ecosystems, communities and
wetlands, how can scientists conduct scienti¢c study
without knowing the exact boundary of their experimental plan? These issues extend to our solar system
and the classi¢cation of planets. What about Pluto?
Is it a planet or not? Current controversy exists over
whether or not Pluto is a planet, a trans-Neptunian
object, a minor, major, or principle planet, a comet,
or an asteroid (McCaughrean et al. 2001; Osorio
2001). Thus, planetary de¢nitions vary widely
depending upon the scientist, the types of features
being considered, the nature of one’s academic background, etc. Stars represent my last example of Individuality beyond species. Stars, like all of the other
examples listed above have their own unique ontogenies that are predictable in many ways because our
knowledge of the physical laws governing the Individuals put into the Natural Kind called Stars is based
essentially on their initial mass. Individual stars are
classi¢ed into Natural Kinds referred to as Red Supergiants, Red Giants, Blue Giants, White Dwarfs and
Main Sequence (Fig. 5). Although all stars did not
evolve from a ‘mother’ star, and not all of the stars of
any one group were derived from a common star of
the group, the famous Hertzprung^Russell classi¢cation derives from a number of qualities of the individual stars such as temperature, absolute magnitude,
and spectral type given o¡ by the star during its ontogeny. Although discrete boundaries between star
types is arguably inconclusive, the evolutionary
ontogeny of a star is predictable. Those stars of the
Main Sequence all have their initial mass roughly
equivalent to 16 suns but they di¡er substantially in
their luminosity, temperature, spectral type and
absolute magnitude. Each of these stars will evolve
with time and some may depart the Main Sequence
and be classi¢ed as Red Supergiants, Giants, or White
Dwarfs. Along the way, however, the star is still the
same Individual governed by natural physical laws
but with boundaries that are less than clear for classi¢cation purposes!
Thus, the issue of Individuality is a much more
dominant occurrence in the natural world than some
may believe.While entities referable to Individuals as
by-products of and governed by natural processes
may be perceived as being di⁄cult to understand,
the bene¢t of making that extra e¡ort to incorporate
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
Figure 5 Planets and stars as
individuals. (A) Example of a star
(Sun) and the traditionally recognized
planets used to illustrate the arti¢cial
construct created by some planetary
scientists delineating Pluto from the
other planets as a nonplanet. (B) The
Hertzprung^Russell diagram and
classi¢cation of stars used to
demonstrate Individuality of stars and
the indistinct boundaries between
Individuals in the Natural Kind star.
this philosophy in to a scienti¢c process results in
promising results that will further our understanding and resolution of the formation and preservation
of our natural world.
special issue is based and the American Society of
Ichthyologists and Herpetologists for developing the
outstanding symposium opportunity to discuss one
of the most important issues in contemporary biology.
Acknowledgements
References
Many people, either in person or through literature,
have been most in£uential in assisting me to formulate my ideas on species. I wish to thank those of you
who have discussed these ideas with me at various
times. In the development of the ideas expressed in
this paper, I speci¢cally wish to thank Darrel Frost,
Anna George, Michael Ghiselin, David Hull, Bernard
Kuhajda, Nick Lang, Michelle Mayden, David Neely,
Steve Powers, Richard Richards, Kevin Roe, Andrew
Simons, E.O.Wiley, Rob Wood and QuentinWheeler. I
also thank the National Science Foundation (USA)
for supporting my systematic studies of ¢shes that
have made my conceptual papers on species, speciation, systematics and evolution possible. Finally, I
thank the organizers of the symposium on which this
Avise, J.C. and Ball Jr, R.M. (1990) Principles of genealogical
concordance in species concepts and biological taxonomy. In: Oxford Surveys in Evolutionary Biology,Vol.7 (eds
D. Futuyma and J. Antonovics). Oxford University Press,
Oxford, UK, pp. 45^67.
Behnke, R.J. (1992) Native Trout of Western North America.
(American Fisheries Society Monograph 6). American
Fisheries Society, Bethesda, MD, USA, 275pp.
Blackwelder, R.E. (1967) Taxonomy: AText and ReferenceBook.
JohnWiley and Sons, NewYork, NY, USA.
Chesser, R.T. and Zink, R.M. (1994) Modes of speciation in
birds: a test of Lynch’s (1989) method. Evolution 48,
490^497.
Coleman, K.A. and Wiley, E.O. (2001) On species individualism: a new defense of the species-as-individuals hypothesis. Philosophy of Science 681, 498^517.
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
193
On species and species concepts R L Mayden
Cracraft, J. (1983) Species concepts and speciation analysis.
In: Current Ornithology, Vol. 1 (ed. R.F. Johnson). Plenum
Press, NewYork, NY, USA, pp.159^187.
Cracraft, J. (1992) The species of the birds-of-paradise (Paradisacidae): applying the phylogenetic species concept to
complex patterns of diversi¢cation. Cladistics 8,1^43.
Cronquist, A. (1978) Once again, what is a species? In: Biosystematics in Agriculture (ed. L.V. Knutson). Allenheld
Osmun, Montclair, NJ, USA, pp.3^20.
Dobzhansky,T. (1950) Mendelian populations and their evolution. American Naturalist 74,312^321.
Dowling,T.E., Smith, G.R. and Brown,W.M. (1989) Reproductive isolation and introgression between Notropis cornutus
and Notropis chrysocephalus (Family Cyprinidae): comparison of morphology, allozymes, and mitochondrial
DNA. Evolution 43,620^634.
Du Rietz, G.E. (1930) The fundamental units of biological
taxonomy. Svensk BotaniskTidskrift 24,333^428.
Eldredge, N. and Cracraft, J. (1980) Phylogenetic Patterns and
the Evolutionary Process: Method and Theory in Comparative Biology. Columbia University Press, New York, NY,
USA.
Frost, D.R. and Hillis, D.M. (1990) Species concepts and practice: herpetological applications. Herpetologica 46, 87^
104.
Frost, D.R. and Kluge, A.G. (1994) A consideration of epistemology in systematic biology with special reference to
species. Cladistics10, 259^294.
Gaston, K.J., ed. (1996a) Biodiversity: a Biology of Numbers
and Di¡erence. Blackwell Science Inc., Malden, MA, USA,
408pp.
Gaston, K.J. (1996b) What is biodiversity? In: Biodiversity: a
Biology of Numbers and Di¡erence (ed. K.J. Gaston).
Blackwell Science Inc., Malden, MA, USA, pp.1^9.
George, T.N. (1956) Biospecies, chronospecies and morphospecies. In: The Species Concept in Paleontology (ed.
P.C. Sylvester-Bradley). London Systematics Association,
London, pp.123^137.
Ghiselin, M.T. (1966) On psychologism in the logic of taxonomic controversies. Systematic Zoology15, 207^215.
Ghiselin, M.T. (1974) A radical solution to the species problem. Systematic Zoology 23,536^544.
Ghiselin, M.T. (1989) Individuality, history, and laws of nature in biology. In: What the Philosophy of Biology Is (ed.
M.J. Ruse). Kluwer Academic Publishers, Dordrecht, pp.
53^66.
Ghiselin, M.T. (1997) Metaphysics and the Origin of Species.
State University of NewYork Press, Albany, NY,377pp.
Grady, J.M. and LeGrande,W.H. (1992) Phylogenetic relationships, modes of speciation, and historical biogeography
of the madtom cat¢shes, genus Noturus Ra¢nesque (Siluriformes: Ictaluridae). In: Systematics, Historical Ecology,
and North American Freshwater Fishes (ed. R.L. Mayden).
Stanford University Press, PaloAlto, CA, USA, pp.747^777.
Hennig, W. (1950) Grundzu«ge Einer Theorie der Phylogentischen Systematik. AufbauVerlag, Berlin, 280pp.
194
Hennig, W. (1966) Phylogenetic Systematics. University of
Illinois Press, Urbana, IL, USA, 280pp.
Highton, R., Maha, G.C. and Maxson, L.R. (1989) Biochemical evolution in the slimy salamanders of the Plethodon
glutinosus complex in eastern US, Illinois. Biological
Monograph 57,1^160.
Howard, D.J. and Berlocher, S.H., eds (1998) Endless Forms:
Species and Speciation. Oxford University Press, New York,
NY, USA.
Hull, D.L. (1976) Are species really individuals? Systematic
Zoology 25,174^191.
Kitcher, P.S. (1984) Species. Philosophy of Science 51,308^333.
Kitcher, P.S. (1987) Ghostly whispers: Mayr, Ghiselin, and
the‘philosophers’on the ontological status of species.Biology and Philosophy 2,184^192.
Kitcher, P.S. (1989) Some puzzles about species. In:What the
Philosophy of Biology Is (ed. M.J. Ruse). Kluwer Academic
Publishers, Dordrecht, pp.183^208.
Kitts, D.B. and Kitts, D.J. (1979) Biological species as natural
kinds. Philosophy of Science 46l,613^622.
Kornet, D. (1993) Permanent splits as speciation events: a
formal reconstruction of the internodal species concept.
Journal ofTheoretical Biology164, 407^435.
Kornet, D.J. and McAllister, J.W. (1993) The Composite Species Concept in Reconstructing Species: Demarcations in
Genealogial Networks. Unpublished PhD Dissertation.
Institute for Theoretical Biology, Leiden, the Netherlands,
pp.61^89.
Kuhn, T.S. (1962) The Structure of Scienti¢c Revolutions. The
University of Chicago Press, Chicago, IL.
Lynch, J.D. (1989) The gauge of speciation: on the frequencies
and modes of speciation. In: Speciation and its Consequences (eds D. Otte and J.A. Endler). Sinauer Press, Sunderland, MA, pp.527^553.
Mallet, J. (1995) A species de¢nition for the modern synthesis. Trends in Ecology and Evolution10, 294^299.
Mallet, J. (1996) The genetics of biological diversity: from
varieties to species. In: Biodiversity: a Biology of Numbers
and Di¡erence (ed. K.J. Gaston). Blackwell Science Inc.,
Malden, MA, USA, pp.13^53.
Mayden, R.L. (1997) A hierarchy of species concepts: the
denouement in the saga of the species problem. In: Species: the Units of Biodiversity (eds M.F. Claridge, H.A.
Dawah and M.R.Wilson). Chapman & Hall Ltd., London,
pp.381^424.
Mayden, R.L. (1999) Consilience and a hierarchy of species
concepts: advances towards closure on the species puzzle.
Journal of Nematology 31,95^116.
Mayden, R.L. and Wood, R.M. (1995) Systematics, species
concepts, and the evolutionarily signi¢cant unit in biodiversity and conservation biology. In: Evolution and
the Aquatic Ecosystem, Special Publication 17 (ed. J.L.
Nielson). American Fisheries Society, Bethesda, MD,
USA, pp.58^113.
Mayr, E. (1940) Speciation phenomena in birds. American
Naturalist 74, 249^278.
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
On species and species concepts R L Mayden
Mayr, E. (1957) Species concepts and de¢nitions. In:The Species Problem (ed. E. Mayr). American Association for
Advancement of Science,Washington, DC, pp.1^22.
Mayr, E. (1969) Principles of Systematic Zoology. McGraw-Hill,
NewYork.
Mayr, E. and Ashlock, P.D. (1991) Principles of Systematic
Zoology. McGraw-Hill, NewYork.
McCaughrean, M., Reid, N.,Tinney, C. et al. (2001) What is a
planet? Science 291,1487.
McKitrick, M.C. and Zink, R.M. (1988) Species concepts in
ornithology. The Condor 90,1^14.
Meier, R. and Willmann, R. (2000) The Hennigian species
concept. In: Species Concepts and Phylogenetic Theory: a
Debate (eds Q.D.Wheeler and R. Meier). Columbia University Press, NewYork, pp.30^43.
Miller, R.R., Williams, J.D. and Williams, J.E. (1989) Extinctions of North American ¢shes during the past century.
Fisheries14, 22^38.
Nixon, K.C. andWheeler, Q.D. (1990) An ampli¢cation of the
phylogenetic species concept. Cladistics 6, 211^223.
Osorio, M.S.Z. (2001) [Response to McCaughrean et al.
(2001) ]. Science 291,1487^1488.
Paterson, H.E.H. (1993) Evolution and the Recognition Concept
of Species: Collected Writings of H.E.H. Paterson (ed. S.F.
McEvey). Johns Hopkins University Press, Baltimore, MD,
USA.
Regan, C.T. (1926) Organic evolution. Report of the British
Association for Advancement of Science1925, pp.75^86.
Ridley, M. (1989) The cladistic solution to the species problem. Biology and Philosophy 4,1^16.
Robins, C.R., Bailey, R.M., Bond, C.E., Brooker, J.R., Lachner,
E.A., Lea, R.N. and Scott,W.B. (1980) Common and Scienti¢c Names of Fishes from the United States and Canada, 4th
edn. Special Publication 12. American Fisheries Society,
Bethesda, MD, USA.
Robins, C.R., Bailey, R.M., Bond, C.E., Brooker, J.R., Lachner,
E.A., Lea, R.N. and Scott,W.B. (1991) Common and Scienti¢c
Names of Fishes from the United States and Canada, 5th
edn. Special Publication 20. American Fisheries Society,
Bethesda, MD, USA.
Roe, K.J. and Lydeard, C. (1998) Species delineation and the
identi¢cation of evolutionary signi¢cant units: lessons
from the freshwater mussel genus Potamilus (Bivalvia:Unionidae). Journal of Shell Fisheries Research17,1359^1363.
Rosen, D.E. (1978) Vicariant patterns and historical explanation in biogeography. Systematic Zoology 27,159^188.
Rosen, D.E. (1979) Fishes from the uplands and intermontane basins of Guatemala: revisionary studies and comparative biogeography. Bulletin of the American Museum
of Natural History162, 267^376.
Shull, G.H. (1923) The species concept from the point of view
fromthegeneticist. AmericanJournalofBotany10,221^228.
Simpson, G.G. (1943) Criteria for genera, species, and subspecies in zoology and paleontology. Annals of the New
York Academy of Science 44,145^178.
Simpson, G.G. (1961) Principles of Animal Taxonomy. Columbia University Press, NewYork, NY, USA.
Sneath, P.H.A. (1976) Phenetic taxonomy at the species level
and above. Taxon 25, 437^450.
Stau¡er Jr, J.R. and McKaye, K.R. (2001) The naming of
cichlids. Journal of Aquariculture and Aquatic Sciences 9,
1^16.
Stau¡er Jr, J.R., McKaye, K.R. and Konings, A.F. (2002) Behaviour: an important diagnostic tool for Lake Malawi
cichlids. Fish and Fisheries 3, 213^214.
Stuessy,T.F. (1990) PlantTaxonomy: the Systematic Evaluation
of Comparative Data. Columbia University Press, NewYork,
NY, USA.
Templeton, A.T. (1989) The meaning of species and speciation: a genetic perspective. In: Speciation and its Consequences (eds D. Otte and J.A. Endler). Sinauer Press,
Sunderland, MA, USA, pp.3^27.
Van Valen, L. (1976) Ecological species, multispecies, and
oaks. Taxon 25, 233^239.
Waples, R.S. (1991) Paci¢c salmon, Oncorhynchus spp. & the
de¢nition of ‘species’ under the Endangered Species Act.
Marine Fisheries Review 53,11^22.
Waples, R.S. (1995) Evolutionary signi¢cant units and the
conservation of biological diversity under the Endangered Species Act. In: Evolution and theAquatic Ecosystem,
Special Publication17 (ed. J.L. Nielson). American Fisheries
Society, Bethesda, MD, USA, pp.8^27.
Warren, M.L. (1992) Variation of the spotted sun¢sh,Lepomis
punctatus complex (Centrarchidae): meristics, morphometrics, pigmentation and species limits. Bulletin of the
Alabama Museum of Natural History12,1^47.
Warren Jr, M.L., Burr, B.M.,Walsh, S.J. et al. (2000) Diversity,
distribution, and conservation status of the native freshwater ¢shes of the southern United States. Fisheries. 25,
7^31.
Wheeler, Q.D. and Meier, R., eds (2000) Species Concepts and
PhylogeneticTheory: A Debate. Columbia University Press,
NewYork, NY, USA.
Wheeler, Q.D. and Platnick, N.I. (2000) The phylogenetic
species concept (sensu Wheeler and Platnick). In: Species
Concepts and Phylogenetic Theory: a Debate (eds Q.D.
Wheeler and R. Meier). Columbia University Press, New
York, NY, USA, pp.55^69.
Wiley, E.O. (1978) The evolutionary species concept reconsidered. Systematic Zoology 27,17^26.
Wiley, E.O. (1981) Phylogenetics: The Theory and Practice of
Phylogenetic Systematics. John Wiley and Sons, NewYork,
NY, USA.
Wiley, E.O. and Mayden, R.L. (2000a) The evolutionary species concept. In: Species Concepts and PhylogeneticTheory:
a Debate (eds Q.D.Wheeler and R. Meier). Columbia University Press, NewYork, NY, USA, pp.70^89.
Wiley, E.O. and Mayden, R.L. (2000b) A critique from the
evolutionary species concept perspective. In: Species Concepts and Phylogenetic Theory: a Debate (eds Q.D. Wheeler
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196
195
On species and species concepts R L Mayden
and R. Meier). Columbia University Press, New York, NY,
USA, pp.146^158.
Wiley, E.O. and Mayden, R.L. (2000c) A defense of the
evolutionary species concept. In: Species Concepts and
Phylogenetic Theory: a Debate (eds Q.D. Wheeler and R.
Meier). Columbia University Press, New York, NY, USA,
pp.198^208.
196
Wilson, E.O. (1992) The Diversity of Life. W.W. Norton, New
York, NY, USA.
Wilson, R.A., ed. (1999) Species: New InterdisciplinaryEssays.
The MIT Press, Cambridge, MA, USA.
Winston, J.E. (1999) Describing Species: Practical Taxonomic
Procedure for Biologists. Columbia University Press, New
York, NY, USA.
# 2002 Blackwell Science Ltd, F I S H and F I S H E R I E S, 3, 171^196